High-availability in multipoint to multipoint Ethernet for the delivery of triple play services
Luis Aguirre-Torres, Gady Rosenfeld · 2006
Ethernet has become a key technology in the development of new end-user applications. One such application is triple play, which offers a service bundle including voice, data and video services. These are offered over the same infrastructure, which means that the service provider needs to take into consideration the combination of the QoS and availability requirements of all individual services combined. A way of offering these services is through the creation of a common broadcast domain through VPLS. This enables multipoint to multipoint connectivity, effectively emulating an extended LAN over the transport infrastructure. While RPR is capable of providing the levels of network availability required to offer triple play services, consistently with customer SLAs, it does not offer any way of protecting against unidirectional physical layer failures in the downstream direction (i.e. the interface between the video server and the local RPR node). VPLS RFI offers a method to propagate a fault indication to all remote nodes part of the VPLS infrastructure. It relies on the generation and propagation of RSVP-TE LOS signaling to all members of the VPLS, resulting in a switchover to an alternate VPLS protection domain. VPLS RFI in combination of RPR, offer high levels of availability in triple play networks, consistent with service level agreements.